Previously the author investigated the relation between air swirl intensity and number of injection nozzle holes by a high speed 2 cycle uniflow scavenging type diesel engine. The fuel injection nozzle was mounted in the center of cylinder head and injected radially. The air swirl was produced by the inclination of scavenging ports. In this paper the anthor studied the relation between the engine performance and the direction of the sprays of 2 fuel injection nozzles mounted on the sides of cylinder head symmetrically. Test engine was the same as used in the previous experiment. The results show that the engine performance was good when the nozzle spray was injected in the same direction as the air swirl. The best angle β (see Fig. 1) of the spray was 30∼40°.

抄録全体を表示

When we use an ordinary carburetter with a cone valve (which is used in the automobile engine), vibrations cause to increase the fuel consumption, and if they are violent they cause to overflow the fuel from float chamber. These are regarded as general phenomena for the system with a float type auto-valve. The experiments have been done chiefly for the cases due to periodic oscillations. These experimental results were compared and examined with the results of calculation which were made under the rough assumptions as a first approximation. As the results of these experiments and calculations the fact that all above-mentioned phenomena come to the vibrating phenomena including the collision of the valve against the valve sheat was clarified, and their mechanisms were resolved clearly. For an example the region where these phenomena are able to occur is suddenly extended when the frequency of oscillation reaches beyund twice as much as the natural frequency of this vibrating system.

抄録全体を表示

In a four cycle engine, if one valve can alternately handle the exhaust and intake air of the cylinder, mixing of which can be prevented by the jet energy of the gas, the following merit may be expected. 1. simplified construction 2. effective valve cooling 3. improvement of volumetric efficiency 4. mixture heating. The experiments were carried out on the same engine alternately with an ordinary valve mechanism and a monovalve one. The performances were compared under the same working conditions. The results of the experiments fairly coincide with the above expectations.

抄録全体を表示

When a fuel droplet is fallen into a properly heated crucible, there exists some time interval between its arrival at the bottom and the initiation of combustion. This is what is called ignition-lag. From the frequency distribution of ignition-lags which were measured by means of a photoelectric-tube, the probability of ignition in unit time at each instant was calculated for different tmperatures of the crucible. It was shown that the ignition-lag consists of two parts, and in the first period the probability of ignition is zero, while in the second period it takes some value. Especially in the case of pure substance this probability becomes constant regardless of time. Furthermore, it was suggested that the first and the second periods correspond to the physical and the chemical delay periods respectively

抄録全体を表示

This is the continued paper of the first report in which the ignition-lag of a single droplet was discussed, and in this paper the ignition-lag of a vapour mass is treated. Using the theory of probability as in the first report, it was pointed out that the ignition-lag of a vapour mass which was suddenly pushed out into the heated air, consists of the physical and the chemical delay periods as that of a single droplet, and that the interdiffusion of fuel vapour and air plays and important part in the physical delay period in this case. We suppose this new knowledge about the ignition-lag of the vapour mass is very available for the comprehension of the ignition-lag of the single droplet and the combustible mixture.

抄録全体を表示

A series of experiments was performed in order to point out the cause of the cycle fluctuation which is unfavorable for the engine operation. As the quantitative expression of the cycle fluctuation, √((x-x)²)^^^-√(x) was calculated from the successive indicator diagrams, where xs are the readings of maximum pressure on the diagrams, and x^^- is their mean value. This quantity was obtained for various operating variables such as mixture ratio, compression ratio, ignition advance, inlet temperature and gap width of the spark plug. By analysing the result of these experiments, we arrived at the following conclusion. There are two important causes of the cycle fluctuation : one is the fluctuation of mixture ratio due to the insufficient vaporization, and the other is the fluctuation of the initial combustion period which is affected by the mixture strength near the electrodes at the moment of spark passing.

抄録全体を表示

From the dimensional analysis for high velocity gas flow in a tube, our recommendable dimensionless modulus, Jλ (Tw-Tg) /3 600 μV², is introduced, which is named "Dissipation Ratio" and expretssd by D_i. In order to recognize the effect of D_i, we performed an experiment, in which air velocity flowing through an annular cross section was 70∼200 m/s, the rate of heat transfered into air from the inner surface 300∼20000 kcal/m²·h and the length of tube 55∼280 mm. From our result, the relation between heat flux and dissipation ratio is obtained and our experimental formulae is expressed by N_u, R_e, L/D and D_i.

抄録全体を表示

The Prandtl-Taylor equation cannot be applied to high velocity gas flow, because the dissipation of friction loss is neglected. In order to obtain the theoretical equation, we make a hypotheses that dissipated heat exists as heat flux from heat source in laminar sublayer, and modify the Prandtl-Taylor equation for high velocity flow. In our formulae, Nusselt number is given by Reynolds number R_e, Prandtl number P_r and dissipation ratio D_i, as in experimental formulae (reference to the 1st and 2nd reports).

抄録全体を表示

In this paper, we studied theoretically the surface heat transmission by natural convection in turbulent boundary layer from the vertical flat plate to the surrounding fluid. For this purpose, we paid attention to the laminar sublayer that exists near the surface and brings very great thermal resistance on the surface heat transmission. Now we assume that in the laminar sublayer, the following relation exists, [numerical formula] where δ_s is the thickness of the laminar sublayer, and T₁, T_s and T₀ are the absolute temperatures of the surface of plate, the surface of laminar sublayer and the surrounding gas, respectively, g is acceleration due to gravity, ν is kinematic viscosity of fluid and P_r is Prandtl number, and G is a constant. Using this assumption as the basis of theoretical analysis, we obtained the next relation among non-dimensional numbers, that is, N_um=0°251P^0.3_rG^0.3_r where N_um is Nusselt number and G_r is Grashoff number. Our theoretical results agree very well with the results which many investigators have already obtained by experiments. And we found that the constant G is about 200.

抄録全体を表示

The velocity distribution w of non-isothermal laminar flow of concentric pipes with annular space has been found and, next, three and two dimensional temperature distributions T necessary for the calculation of heat transfer have been found by expanding the series. On that occasion, boundary conditions f₁ and f₂ are expanded in Fourier Series and, for the sake of simplicity of calculation, assuming, as usual, physical constants to be constant, velocity distribution of isothermal flow and the cylindrical coordinates (r, θ, z) are used.

抄録全体を表示

The method of temperature measurement, utilizing the refraction of light passing through fluid very close to a heated surface, was employed by Schmidt and Saunders (1935-1937). We extended this method and studied the nucleate boiling on horizontal heating surface. When the heating surface sets up normal temperature gradient, the path of light, which has incident angle parallel to the surface, is determined by next equation : (dy/dx)²=2N(θ_b-θ) θ : Temperature at height y above the surface. θ_b : Temperature at height b above the surface. n : Refractive index. N=-(1/n) (dn/dθ) Let Φ denote the maximum inclination of the path of light to the surface after passing through the boundary layer. Φ is measured experimentally. Then, Δθ is obtained using the following relations : Φ²=2NΔθ N=4°36×10^-41/°C (for D-line at 100°C) The values of Δθ measured by optical method showed good agreement with those by thermocouples. Moreover, we devised a simple was of measuring the average Φ varying with time. The effect of the temperature distribution before the edge of the heating surface upon the angle Φ was negligibly small. The light after passing through the layer has the dark and light stripes due to temperature disturbance, and they represent the motion of fluid. The images of bubbles and flow of fluid were ptographed at the same time using revolving drum. At lower heat flux these stripes move uniformly, showing the laminar motion of the temperature disturbances, and at higher heat flux the flow of motion becomes turbulent. The thickness of temperature boundary layer could be also measured easily. The curve of thickness of boundary layer plotted vs, coefficient of heat transfer has a bending point, which shows the difference between laminar heat transfer and turbulent heat transfer. This agrees with the results of measuring of motion in the boundary layer.

抄録全体を表示

In order to clarify the mechanism of heat transfer in nucleate boiling, we have taken many photographs of generation of vapour bubbles on a horizontal heated plate (dia. 10 cm) and fluctuation of boundary layer by utilizing the refraction of light passing through the distilled water very close to a heated surface and at the same time we have measured the temperature of a heated plate and the temperature distribution from the surface of plate to the water level by thermocouples or an optical method (see 1 st report). The range of heat flux is from 6560 to 43760 kcal/m²h. The conclusions are shown as follows : (1) The flows in the boundary layer close to a heated surface vary as the type of generating vapour bubbles. (2) Heat flux below 9800 kcal/m²h corresponds to the free convection region where stirring effect of steam bubbles is little and heat flux over 15000 kcal/m²h the forced convection region where stirring effect is important, and between them there is the transition region from free convection to forced convection. (3) In the heat transfer to boiling liquid, the important factor controlling the coefficient of heat transfer is not the heat flux or the temperature difference between the surface of a heated plate and boiling liquid, but the stirring effect of vapour bubbles. (4) No relating to type, size and frequency of generating steam bubbles, we obtained the following relations in the first approximation. α∞ω^0.5 and α∝η^1/3 where α=coefficient of heat transfer to boiling water. w=average velocity of temperature boundary layer close to a heated plate. n=number of vapour columns on a heated plate.

抄録全体を表示

Even if the exact solution of the conduction of heat is obtained by the analytical method, the numerical caluculations are laborious, especially in the case of radial flow of heat. The author has studied Takahashis' method and has succeeded to extend his method tot he cases of the radial flow of heat, i.e. temperature distribution of the infinite solid with a circular hollow, where the temperature is constant or varies as time.

抄録全体を表示

As to the brake of rotating machines, it must be sure that the work done in braking action transforms into heat energy. Therefore, if the generated heat is dispersed by temperature rising of brake and reflection from the outer surface, this phenomenon is a problem of heat conduction in solid bodies. By way of the severe example, heating of automobile brakes is the most suitable. And in the case of this calculation, since the ratio (radius and thickness of brake) is probably small, the effect of curvature and edge effect may be ignored and the equation for plane faces used. According to a numerical evaluation it has been found that the rising tendency in temperature of the working surface is extremely high when the car stops.

抄録全体を表示

There are a number of problems in heat conduction in which one material is transformed in to another with generation or absorption of heat. Examples are the melting or freezing of a solid and the evaporation of a liquid droplet. Generally speaking, these problems can not he solved strictly by analytical methods. We could obtain a general solution by application of Takahashis' method. As the first example, the melting velocity and temperature distribution of a metal were calculated and the results were shown.

抄録全体を表示

When a liquid droplet is injected into high temperature gas by an injection valve, at once the liquid droplet begins to evaporate and conducts heat. These problems can not be solved strictly by analytical methods. We solved this problem by application of Takahashis' method. As the examples, the evaporation velocity and temperature distribution of a liquid droplets of water and benzol were calculated and the results were showen.

抄録全体を表示

The infra-redray was projected on the test piece, and the dryness of paint was observed. Further I studied the dry velocity and evaporated water and the amount of contained water etc. A new attempt was executed about the observation on the change of temperature at the inner part of the paint.

抄録全体を表示

In the 2nd. report the rerearch on the character of infra-red drying was discussed. In this paper, the mechanism of infra-red drying of "clay" is reported, being compared with convectiondrying. The critical moisture-content is lower in the former than in the latter drying. The drying process resembles to convection drying against high frequecy wave drying, the gradient of moisture-content towads the outer surface being very gentle. Consequenty, in drying of clay, the skin effect (that only very thin outer layer is dried) can be avoided. In drying of clay from 65% to 14% (in dry basis percentage), 1/2 in electric power and 2/5 in necessary time are seconsumedved comparing with convection-drying.

抄録全体を表示

Combining a refrigerating cycle by a vapor-compressor and an air cycle by a blower, we get a new-type dryer, in which, circulating air is dehumidified by the evaporator of the refrigerator, and then heated by the condenser. This dry air is utilized in the drying process. We deal theoritically with the several cycles concerning this dryer, the necessary conditions to keep the cycles in steady states, and relative merits of the cycles. We are able to expect the following advantages by this dryer. (1) The high heat efficiency (2) To control both temperature and humidity separately as required. (3) The drying in low temperature economically. We experimented with this type of dryer by the methyl-chloride refrigerator, and then obtained some conclusions about its design.

抄録全体を表示

There are two representative methods to investigate the characters of turbine nozzles. They are "impulse" and "reaction" methods. Originally the results with these two methods must agree with each other, but practically they do not always. May we deduce the characters of air or gas turbine nozzles from the known characters of steam turbine nozzles? To get answers to these points, we have improved the impulse method and researched the characters of turbine nozzles using air, and compaired the results with that which I got before with reaction method using steam. The results of researches with impulse method using air coincide generally with those of reaction method using steam. For example the mean values of velocity coefficients are 0.94∼0.95.

抄録全体を表示

The radial flow air impulse turbine has often been used to obtain high speed. Previously the author presented some data for designing this type of turbine. According to these data, a well designed turbine has been used to drive our "Ultra-centrifuge". Then we are able to obtain the maximum speed of 60000 rpm and to satisfy the required performance. In this paper, the results obtained in the actual running and heat balance of the Ultra-centrifuge are shown.

抄録全体を表示

In this report, the following matters are explained from the result of our boiler test : 1. As the efficiency-load curves which have hitherto been used as the characteristic curves are not satisfactory, we recommend to use efficiency-load-excess air ratio surface. 2. By our testing method, we can accomplish the boiler test at high accuracy within about an hour, including the pre-operation to uniformize the combustion. 3. Boiler efficiency is a function of load, excess air ratio, draft, temperature of the flue gas, damper opening etc. two of which are independent variables. Hence we can operate the boiler at the state of maximum efficiency for a given load, when we observe any one of the other variables for example draft or damper opening.

抄録全体を表示

As the first stage of study on the steam ejectors, it is the purpose of this paper to know experimentally the condition of flow and the extent of mixing of steam and inhaled gas (air on this experiment) at each cross-section of the ejector. By the experiment we have obtained the characteristic curves, the relation of vacuum of mixing chamber and weight of inhaled air. And on each characteristic curve, we have found the condition and extent of mixing of flow by means of measuring the static and Pitot pressures on each section of the diffuser. By these results, we have made clear the so-called hysteresis phenomena of steam ejector in which two characteristic curves are present on the same outside conditions, high vacuum and low vacuum characteristics, and shown the difference of the two characteristics.

抄録全体を表示

The flow in diffuser of the steam ejectors consists of the two processes, one of which is to mix steam with inhaled gas and the other is to transform the energy from the velocity of flow into the pressure. But in the theories of the steam ejectors, usually, the mixing process is. neglected, so that these theoretical methods are, as pointed out by Dr. Stodola, of little value for the ditermination of the dimensions in an actual case for the following reason : We do not know to what extent the velocities in a cross-section are equalized. Consequently, neither an exact energy equation nor a continuity equation can be formulated. So the author tried a calculation method of the flow in diffuser of the steam ejectors by thermodynamic consideration in which the two processes, not only the energy transformation process, but the mixing process, are considered. And he has made clear the various changes of flow state in diffuser and the causes of the so-called hysteresis phenomena of steam ejector. By this consideration, the two required conditions have been obtained for high vacuum in mixing chamber. (high vacuum characteristic) On the thermodynamic method, we meet with a difficulty, as also pointed out by Dr. Stodola, that the flows are of different kinds (steam and air). But this difficulty has been solved in this paper by drawing up i, s-diagrams for each air ratio, so that the various changes of flow state are shown on these diagrams.

抄録全体を表示